Iris Image Classification Based on Hierarchical Visual Codebook,

Zhenan Sun, Hui Zhang, Tieniu Tan, Jianyu Wang

Abstract: Iris recognition as a reliable method for personal identification has many important applications in both public and personal security areas. Iris recognition has been well studied with the objective to assign class label of each iris image to a unique subject. In contrast, iris image classification aims to classify an iris image to an application specific category, e.g. iris liveness detection (classification of genuine and fake iris images), race classification (classification of iris images of Asian and non-Asian subjects), coarse-to-fine iris identification (classification of all iris images in the central database into multiple categories). This paper proposes a general framework for iris image classification based on texture analysis. A novel texture representation method called Hierarchical Visual Codebook (HVC) is proposed to encode the texture primitives of iris images. The HVC is an integration of two existing Bag-of-Words models, namely Vocabulary Tree (VT) and Locality-constrained Linear Coding (LLC). It adopts coarse-to-fine visual coding strategy like VT and sparse representation of LLC. Extensive experiments demonstrate that the proposed method achieves state-of-the-art performance for iris liveness detection, race classification and coarse-to-fine iris identification. A comprehensive fake iris image database simulating four types of iris spoof attacks is developed as the benchmark for research.